Planetary and Space Science - UChicago...

Planetary and Space Science 181 (2020) 104820

Contents lists available at ScienceDirect

Planetary and Space Science

journal homepage: www.elsevier.com/locate/pss

Methane release on Early Mars by atmospheric collapse andatmospheric reinflation

Edwin S. Kite a,*, Michael A. Mischna b, Peter Gao c,d, Yuk L. Yung b,e, Martin Turbet f,g,1

a University of Chicago, Chicago, IL, 6063, USAb Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, 91109, USAc NASA Ames Research Center, Mountain View, CA, 94035, USAd University of California, Berkeley, CA, 94720, USAe California Institute of Technology, Pasadena, CA, 91125, USAf Laboratoire de M�et�eorologie Dynamique, IPSL, Sorbonne Universit�es, UPMC Univ Paris 06, CNRS, Paris, 75005, Franceg Observatoire Astronomique de l’Universit�e de Gen�eve, 51 Chemin des Maillettes, 1290, Sauverny, Switzerland

A B S T R A C T

A candidate explanation for Early Mars rivers is atmospheric warming due to surface release of H2 or CH4 gas. However, it remains unknown how much gas could bereleased in a single event. We model the CH4 release by one mechanism for rapid release of CH4 from clathrate. By modeling how CH4-clathrate release is affected bychanges in Mars’ obliquity and atmospheric composition, we find that a large fraction of total outgassing from CH4 clathrate occurs following Mars’ first prolongedatmospheric collapse. This atmosphere-collapse-initiated CH4-release mechanism has three stages. (1) Rapid collapse of Early Mars’ carbon dioxide atmosphereinitiates a slower shift of water ice from high ground to the poles. (2) Upon subsequent CO2-atmosphere re-inflation and CO2-greenhouse warming, low-latitudeclathrate decomposes and releases methane gas. (3) Methane can then perturb atmospheric chemistry and surface temperature, until photochemical processesdestroy the methane.

Within our model, we find that under some circumstances a Titan-like haze layer would be expected to form, consistent with transient deposition of abundantcomplex abiotic organic matter on the Early Mars surface. We also find that this CH4-release mechanism can warm Early Mars, but special circumstances are requiredin order to uncork 1017 kg of CH4, the minimum needed for strong warming. Specifically, strong warming only occurs when the fraction of the hydrate stability zonethat is initially occupied by clathrate exceeds 10%, and when Mars’ first prolonged atmospheric collapse occurs for atmospheric pressure >1 bar.

1. Introduction

1.1. Causes and effects of methane release on Early Mars

The Mars Science Laboratory’s (MSL’s) investigation of Early Marssediments has turned up two results which remain a puzzle. The firstpuzzle is the discovery of a paleolake (Grotzinger et al. 2014). Paleolakeson Mars individually lasted as long as >(102–103) yr (e.g., Irwin et al.,2015). Such longevity is hard to reconcile with climate models thatpredict no lakes, or only short-lived lakes (Wordsworth, 2016; Hynek,2016; Luo et al., 2017; Haberle et al., 2017; Vasavada, 2017; Kite et al.,2019; Kite, 2019). The second puzzle is the detection by MSL of organicmatter (Eigenbrode et al., 2018). The organic matter is found in excess ofknown instrumental backgrounds. The source could be exogenic (aster-oidal dust), or indigenous to Mars (for example, magmatic or biological),but remains unknown.

* Corresponding author.E-mail address: [email protected] (E.S. Kite).

1 Currently Marie Sklodowska-Curie Fellow, Geneva Astronomical Observatory.

https://doi.org/10.1016/j.pss.2019.104820Received 30 August 2019; Received in revised form 4 December 2019; Accepted 5 DAvailable online 2 January 20200032-0633/© 2020 Elsevier Ltd. All rights reserved.

Both these puzzles – lake-forming climates and organic matter pro-duction – could, in principle, be explained by CH4 release. Methane is agreenhouse gas. For one-dimensional calculations of Early Mars climate,CH4–induced warming (due to CH4–CO2 collision-induced absorption)overwhelms CH4–induced cooling (due to CH4 absorption of sunlight)when pCO2 exceeds 0.5–1.0 bar (Wordsworth et al., 2017; Turbet et al.,2019). Indeed, CH4 bursts have been proposed as a cause of early Marslake-forming climates (Chassefi�ere et al., 2016; Kite et al., 2017a), and asa cause of various Early Mars geologic features (e.g., Skinner and Tanaka,2007; Kite et al., 2007; Oehler and Allen, 2010; Wray and Ehlmann,2011; Komatsu et al., 2016; Ivanov et al., 2014; Oehler and Etiope, 2017;Pan and Ehlmann, 2014; Etiope and Oehler, 2019; Bro�z et al., 2019).Detections of tiny amounts of CH4 in Mars’ present-day atmosphere havebeen reported (e.g. Webster et al., 2018), although these detections donot all agree with one another and are in tension with models that predictlower methane abundance, and also lower methane abundance

ecember 2019

mailto:[email protected]

http://crossmark.crossref.org/dialog/?doi=10.1016/j.pss.2019.104820&domain=pdf

www.sciencedirect.com/science/journal/00320633

http://www.elsevier.com/locate/pss

https://doi.org/10.1016/j.pss.2019.104820


E.S. Kite et al. Planetary and Space Science 181 (2020) 104820

variability, than has been reported (Zahnle et al., 2011; Mischna et al.,2011; Korablev et al., 2019; Zahnle and Catling, 2019; Moores et al.,2019). Separate from its potential to warm Mars, methane can also serveas the feedstock for photochemical production of complex organic matter– hazes and soots. Haze-derived soot on Titan today, which is ultimatelyderived from CH4, accumulates on Titan’s surface (Lunine and Atreya,2008; H€orst, 2017). A ratio of CH4/CO2> 0.1 is needed for vigorous hazeproduction (Trainer et al., 2006; Haqq-Misra et al., 2008; Kite et al.,2017a).

Here we investigate a new scenario for Mars methane release (Fig. 2).The scenario involves both collapse of the CO2 atmosphere, and also thereverse process – CO2-atmosphere reinflation (Gierasch and Toon, 1973;Soto et al., 2015; Wordsworth et al., 2015). Atmospheric collapse onMars shifts the climate from a state with all CO2 in the atmosphere, to astate with most CO2 in the form of surface ice, and atmospheric reinfla-tion is the reverse process. In our model, atmospheric-collapse-triggereddepressurization of CH4 clathrate and subsequent reinflation-inducedwarming of CH4 clathrate causes transient outgassing of CH4.

In order to model these processes, our approach combines thefollowing elements (see Appendix A for details): climate models, obliq-uity calculations, a parameterization of the photochemical destruction ofCH4 that is fit to a detailed photochemical model, and a model of the(latitude/longitude/depth-dependent) stability of CH4 clathrate. Forsimplicity we omit other clathrates such as CO2 clathrate (and CH4–CO2clathrate hybridization; Sloan and Koh, 2008). We emphasise CH4clathrate for the following reasons. Methane clathrates are a candidatedriver of past climate change on Titan and Earth (e.g., Tobie et al., 2006;Bowen et al., 2016). Moreover, CH4 storage in clathrates has been pre-viously proposed for Mars (Chassefi�ere et al., 2016, and referencestherein), and as a possible source of the CH4 reported by a team using

Fig. 1. This paper in context: candidate drivers and sources for transient release of Hunderway to determine the conditions (if any) under which transient reducing greenmechanism. One chemical reaction that could link candidate driver a to transient releH2O → FeO þ H2). Figure notes: a: e.g. Haberle et al.2008, Steakley et al. 2019. b: eRamirez 2017. d: Tosca et al. 2018. e: e.g. Chass�efiere et al. 2016. f: eg. Lin et al. 2

2

MSL Sample Analysis at Mars (SAM) Tunable Laser Spectrometer (TLS)data (Webster et al., 2018; but see also Zahnle et al., 2011; Zahnle andCatling, 2019). However, CH4 has a <106 yr photochemical lifetime(Sagan, 1977; Gierasch and Toon, 1973; Chassefi�ere et al., 2016; Lasueet al., 2015; Wordsworth et al., 2017; Kite et al., 2017a). In order to buildup the high (�1%) levels of atmospheric CH4 that are needed for strongwarming on Early Mars, CH4 supply must overwhelm photolysis, and sooutgassing must be swift. In turn, swift outgassing requires a mechanismthat (a) efficiently traps CH4, yet (b) subsequently releases the CH4, (c) inlarge quantities (d) at a rate that outpaces photochemical destruction.Here, we show that a suitably vigorous release mechanism is atmosphericcollapse and atmospheric reinflation.

1.2. Previous work on Mars atmospheric collapse and atmosphericreinflation

CO2-atmosphere collapse (a runaway shift from intermediate to verylow PCO2) is distinct from CO2-atmosphere condensation (which curtailsPCO2 at high pressure, ≳3 bars, without involving a runaway shift). Themost recent 3D analysis of runaway atmospheric collapse on Mars is bySoto et al. (2015). In their study, which assumed present-day solar lu-minosity, collapse onto Olympus Mons plays a key role. However, thisvolcano was probably not very tall at 3.7 Ga (Isherwood et al., 2013).Idealized models have also been used to study atmospheric collapse onMars (Nakamura and Tajika, 2001, 2002, 2003), and on exoplanets (e.g.,Heng and Kopparla, 2012; Wordsworth, 2015; Turbet et al., 2017). Allmodels agree that, for a given PCO2, the main control on atmosphericcollapse is obliquity. The obliquity for atmospheric collapse depends onassumed CO2 ice cloud grain size (Kitzmann, 2016). Using a 3D model,Kahre et al. (2013) assessed the importance of carbon dioxide ice cap

2 or CH4 gas on Early Mars. A major effort, involving many research groups, ishouse atmospheres on Mars are geologically plausible, by testing each proposedase of H2 is reaction between Fe from the impactor and H2O from Mars (i.e., Fe þ.g., Prieto-Ballesterous et al. 2006, Kite et al. 2017a. c: e.g., Batalha et al., 2016,005, Tarnas et al. 2018

Fig. 2. Overview of the collapse-initiated methane release mechanism. Upper panel: Depth-versus-time schematic of the long-term evolution of a column of the low-latitude highlands of Mars. Figs. 3–5 show details. Lower panel: Latitude-versus-time schematic, showing key volatile reservoirs: CO2, H2O, and CH4. Reservoirthicknesses are not to scale. Below the critical obliquity for atmospheric collapse (φ c), 90–99.9% of the atmosphere will condense in ~1 kyr (Step 1). This unloadshigh ground (Step 2), releasing some CH4 from sub-ice clathrate. Later re-inflation of the atmosphere triggered by φ rise (Step 3) leads to massive CH4 release andpossible climate warming (Step 4). The basis for the timescale estimates is GCM output, as described in the text.


albedo and emissivity in setting the boundaries of the atmosphericcollapse zone. Manning et al. (2006, 2019) considered atmosphericcollapse as part of a study of the long-term evolution of the Mars carboninventory. There is strong geologic evidence for major changes in PCO2

2 New radar estimates on the present-day inventory of CO2 ice (Bierson et al.,2016; Putzig et al., 2018) mean that warming driven by the direct greenhouseeffect of CO2 released into the atmosphere during re-inflation in the geologicallyrecent past or near future is probably small (Jakosky and Edwards, 2018). This iscontrary to the large warming envisaged by early studies (Sagan et al., 1973;Gierasch and Toon, 1973; McKay et al., 1991). Nevertheless, earlier in Marshistory, Mars had more CO2, and so the warming driven by the direct green-house effect of CO2 released into the atmosphere during re-inflation early inMars history could be much larger (e.g., the effect of going from 0.01 → 1 barPCO2), as is assumed in this study.

3

due to redistribution of CO2 between the atmosphere and CO2 ice capseven in the geologically recent past, within the last 5 Myr (e.g., Kre-slavsky and Head, 2005; Phillips et al., 2011; Bierson et al., 2016;Manning et al., 2019).2 No previous study links Mars atmosphericcollapse and reinflation to a warmer-than-pre-collapse climate (Halevyand Head, 2017).

2. Setting the stage for atmospheric-collapse-initiated CH4release

The atmospheric-collapse-initiated methane release scenario ismotivated by evidence for the following:- (1) Noachian-age water-rockreactions in the deep subsurface; (2) a ≳0.5 bar past atmosphere that (3)drove surface H2O ice to high ground; and (4) large-amplitude obliquityvariations. Understanding of all four has improved in recent years, as we

Fig. 3. Sketch of atmospheric collapse on Mars, showing how atmosphericcollapse drives H2O-ice re-distribution. Thin black lines show annual-mean polartemperature as a function of atmospheric pressure assuming Faint Young Suninsolation (75% of modern insolation). These lines can shift depending on modelassumptions, and the lines shown here are illustrative only. Thick black line isthe condensation curve for CO2; atmospheres below this line are collapsing ontopolar CO2 ice caps (e.g., A→B). Blue dashes outline the approximate pressuresand obliquities below which surface H2O ice is stable only at Mars’ poles (e.g.,Mischna et al., 2013; Wordsworth, 2016). Collapse drives a shift of surface H2Oice from highlands to poles. In this sketch, for an initial CO2 inventory of 8 �1018 kg (¼ 2 bar), the atmosphere is stable until obliquity (φ) � 15� (at A).Rapid collapse (~103 yr) moves the system to a very low atmospheric pressure(lower than point B). Increasing obliquity (over 105–107 yr) moves the (icecap)/atmosphere system along the condensation curve to C, (the highest φ

consistent with permanent CO2 ice caps). Further φ rise leads to sublimation ofthe CO2 ice cap (~103 yr), and the system returns to A.

Fig. 4. CH4 clathrate phase diagram. Phase boundaries shown in black. Thetemperatures are obtained from our GCM (which is modified after Mischnaet al., 2013; see Appendix for details). Mars geotherms are shown in by uppercurved line (early, steep geotherm; red) and lower curved line (later, shallowgeotherm; orange). Early in Mars history, near-surface cooling locks-in CH4 asclathrate in regolith beneath ice sheets (dashed gray line with arrow). Furthergeotherm cooling and escape of ice-sheet H2O to space has little effect onCH4-clathrate stability. Atmospheric collapse and consequent unloading causeminor CH4-clathrate breakdown (Fig. 2) (Steps 1–2). Warming of the surfaceupon re-inflation, plus feedback warming, will cause major CH4-clathratebreakdown (Steps 3–4).


now discuss.

1) Noachian-age hydrothermal minerals record Noachian hydrothermalreactions (e.g., Carter et al., 2013; Ehlmann et al., 2010; Parmentierand Zuber, 2007; Sun and Milliken, 2015). Hydrothermal reactionsbetween Mars’ mafic/ultramafic crust and waters charged withmagmatic and/or atmospheric C should yield both H2 and CH4 (e.g.,Lyons et al., 2005; Oze and Sharma, 2005; Klein et al., 2019). CH4 isemphasized here (Etiope and Sherwood Lollar, 2013), due to its sta-bility in clathrate hydrates under lower pressures and higher tem-peratures relative to H2. Although abiotic reactions can, in principle(stoichiometry þ thermodynamics), form ~102 bars of CH4, CH4

production by abiotic reactions faces kinetic barriers (e.g., Oze et al.,2012; Tarnas et al., 2019). These barriers can be overcome by hightemperatures in olivine-rich rocks that host catalysts (e.g., Ni; Etiopeand Schoell, 2014), and by recycling of fluid by hydrothermal cir-culation. Both hydrothermal circulation, the hydrothermalreaction-rate, and radiolytic H2 production slackened as the Marsgeotherm cooled and Mars’ radionuclide abundances decreasedthrough radioactive decay (Ehlmann et al., 2011; Fassett and Head,2011; Tarnas et al., 2018). Hydrothermal circulation would transportCH4 that was produced in the deep subsurface up to the coolingnear-surface (e.g., beneath ice sheets or primordial seas), where itwould have been trapped as CH4 clathrate hydrate (Fig. 2) (Chastainand Chevrier, 2007; Chassefi�ere and Leblanc, 2011; Mousis et al.,2013). CH4-clathrate is not stable at Mars’ surface (Fig. 4), but isstable in the subsurface. Specifically, CH4-clathrate is stabilized byincreasing pressure (stable for P � 21 bar at 273.15 K, correspondingto a depth of 400 m within 1.5 g/cc regolith), and by decreasingtemperature (stable for P � 0.015 bar at 200 K) (Fig. 4).

4

Shallow-subsurface CH4 clathrate, once formed, could later bedestabilized by orbital forcing (Kite et al., 2017a; Prieto-Ballesteroset al., 2006) or by other mechanisms (Chassefi�ere et al., 2016; thispaper) (Fig. 1).

2) Today, PCO2 ¼ 6 mbar, too low to permit extensive liquid water(Hecht, 2002). Past PCO2 was greater, due to gradual atmospheric lossby hard-to-reverse processes (carbonate formation, polar basalmelting and infiltration of CO2) as well as by irreversible escape tospace (Kurahashi-Nakamura & Tajika 2006, Lammer et al., 2013;Jakosky et al., 2018; Manning et al., 2019). Most estimates of PCO2 atthe time of the valley networks are in the range 0.2–2 bar (Kite et al.,2014; Kurokawa et al., 2017; Cassata et al., 2012; Warren et al., 2019;Kite, 2019 and references therein). Such thick CO2 atmospheresensure that cold ground is at high elevations (Wordsworth et al. 2013,2015).

3) Cold ground at high elevations, including most of the Southernhemisphere, would act as a cold trap for H2O ice (Wordsworth et al.,2013). Average H2O ice sheet thickness, assuming ice above þ1 kmelevation, was �300 m (e.g., Fastook and Head, 2015; Mahaffy et al.,2015). Three hundred meters of H2O ice on Mars is sufficient tostabilize CH4 clathrate in the regolith pore space beneath the icesheet.

4) Unlike Earth, Mars is thought to undergo chaotic large-amplitudeshifts in spin-orbit parameters; among these variations, the mostimportant are variations in obliquity, φ. φ varies in the range 0� < φ< 70�; Touma and Wisdom (1993), Laskar et al., (2004), Holo et al.,(2018), on which are superposed quasi-periodic φ variations withperiod 105–106 yr and amplitude 0–20� (10 � Earth).

3. How atmospheric collapse can initiate methane release

Here, we summarize the collapse-initiated CH4-release scenario (amore detailed description is given in x3.1). In the atmospheric-collapse-

Fig. 5. (a) Map of initial depth (within regolith), in meters, to the top of the CH4-clathrate hydrate stability zone (HSZ). This depth is zero where H2O ice (located attopographic elevations >þ1 km, following Fastook and Head, 2015) is thick enough to stabilize CH4 clathrate throughout the regolith. Vertical line at 27�E is shown incross-section (b). (b) Latitude-depth cross-section showing the depth to the top of the CH4-clathrate hydrate stability zone. Blue line corresponds to initial HSZboundary. Red line corresponds to cold, post-collapse conditions, assuming complete H2O movement to the poles. Yellow line corresponds to re-warmed, post--reinflation conditions, assuming H2O has not yet returned from the poles, but not including CH4 warming. Green line shows effect of 10 K of further warming (e.g., dueto CH4–CO2 CIA).


initiated CH4-release scenario (Fig. 2), atmospheric collapse (Step 1)causes low-latitude surface water ice to sublimate away. This sublimationloss occurs because atmospheric collapse occurs at low obliquity. At lowobliquity and low (collapsed) pCO2, the stable location for water ice is atthe poles (as on Mars today). Loss of water ice overburden depressurizesand thus destabilizes CH4 clathrate in low-latitude sub-glacial pore space(Step 2). Subsequent atmospheric re-inflation at high obliquity leads toCO2-induced warming that further destabilizes CH4 clathrate (Steps 3/4).If (and only if) the atmospheric collapse is initiated at pCO2 > (0.5–1.0)bar, then CH4 release can cause surface warming. The potential for sur-face warming is greatest at Step 4. The CH4 pulse is brought to a close byphotochemical destruction of CH4. We proceed through each of thesesteps in greater detail in x3.1.

The collapse-initiated CH4 release mechanism (Fig. 2) is motivated bytwo separations of timescales:

1. CO2-atmosphere collapse and re-inflation takes ~103 yr (Soto et al.,2015). This ~1 mbar/yr pace is set by the need to radiate away (orsupply) the latent heat of sublimation for CO2 ice. Both CO2 collapseand CO2migration are slow relative to H2O ice migration (105–107 yr,according to our GCM; Appendix). (H2O ice migration is slow becausesurface water ice is cold on Early Mars, cold ice has a low saturationvapor pressure, and as a result there is not much water vapor in theatmosphere for winds to transport). Since H2O ice overburden pres-sure is important to CH4 clathrate stability, this separation of time-scales allows for H2O ice to be “out of position”when the atmospherere-inflates. As a result, after reinflation, CH4 clathrate is at the sametemperature, but a lower pressure, than in thepre-atmospheric-collapse state. Because CH4 clathrate is destabilizedat lower pressure (and destabilized at increasing temperature), CH4clathrate is destabilized after reinflation relative to pre-collapseconditions.

2. The time lag between surface warming and release of CH4 to the at-mosphere is the sum of time for the subsurface to be warmed bythermal conduction, the decomposition time for unstable clathrate(which is fast; Gainey and Elwood Madden, 2012), and the time lagbetween clathrate breakdown and release of CH4 to the atmosphere.This sums to ≲103 yr, much less than the timescale for photochemical

5

destruction of CH4 (~105 yr). Thus, CH4 can build up to high levelsbefore decaying.

3.1. Steps in the scenario

Step 1. Collapse of an initially thick CO2 atmosphere. We assume aninitially thick (PCO2 > 1 bar) atmosphere. As the atmosphere is slowlythinned by escape to space and other processes, the remaining atmo-spheric CO2 becomes increasingly vulnerable to collapse (Fig. 3).Collapse is triggered if polar CO2 ice caps undergo year-on-year growth.Growth of polar CO2 ice caps occurs below a critical polar temperature.Polar temperature is set by insolation, which increases with obliquity,and by heating from the CO2 atmosphere (both the greenhouse effect andequator-to-pole heat transport increase with PCO2). Given secularatmospheric-pressure decline, obliquity variations will eventually lowerinsolation below the threshold for perennial CO2-ice caps (e.g., Forgetet al., 2013). Once year-on-year cap growth begins, the caps trap 90–99%of the atmospheric CO2 within 103–104 yrs (Soto et al., 2015). Why is thiscollapse so rapid, and (almost) complete? The cause of the rapid runawayis the positive feedback between polar cooling and ice-cap sequestrationof CO2. Post-collapse, PCO2 is in equilibrium with polar CO2 ice capsurface temperature - i.e., it is very low (Fig. 3) (Sagan et al., 1973). CO2ice cap thickness, assuming caps poleward of 80�, is ~1200 m/bar CO2(or lower if the caps flow; Mellon, 1996).

Low-obliquity atmospheric collapse involves a hysteresis (Fig. 3,Fig. A2), but is ultimately reversible. Eventually, obliquity rise willtrigger rapid re-sublimation of CO2 condensed at low obliquity. Thus,CO2 trapped as CO2 ice is only sequestered temporarily.

Theory predicts that collapse/re-inflation cycles should haveoccurred in Mars history. There is sedimentological evidence for PCO2 <10 mbar at ~3.7 Ga (Lapotre et al., 2016), in which case the first collapseoccurred >3.7 Ga. There is also indirect evidence from isotopes in Marsmeteorite ALH 84001 that collapse (i.e, a very thin atmosphere) did notoccur prior to ~4 Ga (Kurokawa et al., 2017).

PCO2 reduction leads to loss of most of the greenhouse effect. Thecorresponding surface cooling (tens of K) helps to stabilize CH4-clathrate,

Fig. 6. Example climate evolution model output (for f ¼ 0.15; sensitivity to f is shown in Fig. A4). This is a speculative calculation because it assumes atmosphericcollapse is initiated at >0.8 bar pCO2, higher than predicted by currently-available models (Appendix A). Top: Pressure evolution. Partial pressure of CO2 (blue) andCH4 (red). Bottom: Temperature evolution. After re-inflation, a ~100 kyr-long, >10 K warming occurs. Green highlights the >104-yr long intervals when annual meantemperatures in the �40� latitude, �2 to þ3 km elevation zone (valley network zone; Hynek et al., 2010) exceed those for lakeshore weather stations in Taylor Valley,Antarctica (Doran et al., 2002). This calculation is for the CH4-induced warming of Turbet et al. (2019); similar calculations using the parameterization of Wordsworthet al. (2017) lead to longer, more intense warm periods.


far outweighing the destabilizing effect of loss of the weight of the at-mosphere. Therefore, no CH4 release occurs at Step 1. Furthermore, coldpost-collapse conditions are very unfavorable for liquid H2O on EarlyMars (Hecht, 2002; Turbet et al., 2017).

Step 2. H2O ice unloading of low latitudes triggers some CH4 release.Following atmospheric collapse, the collapsed <100 mb atm can nolonger supply much heat to the poles. Lacking atmospheric heating, andwith obliquity low in the aftermath of atmospheric collapse, the coldpoles are now the stable cold-trap location for water ice. Water ice con-denses at the poles, and sublimates away from low-latitude highlands.The sublimation rate is slow, due to the Faint Young Sun: ~0.1 mm/yr forwater ice albedo ¼ 0.45 (according to our general circulation model,which is a modified version of that in Mischna et al., 2013; Appendix A),rising to ~1 mm/yr for a dust-like ice albedo. Thus, the sublimationtimescale for removal of the low latitude H2O-ice depends on the initialH2O thickness. For a thickness of 300 m, removal/unloading time is 3 �105-3 � 106 yr. On this timescale the absolute distance traveled by H2Oice via solid-state glacial flow is small (Kite and Hindmarsh, 2007; Fas-took and Head, 2015).

This slow latitudinal shift in H2O ice overburden pressure de-pressurizes and thus destabilizes CH4 clathrate in subglacial pore space(Figs. 4-5). Unloading-induced destabilization eventually exceedscooling-induced stabilization. In other words, the top of the clathratehydrate stability zone (HSZ) will deepen, so the hydrate stability zonewill shrink. The deepening HSZ boundary will cause clathrate strandedabove it to decompose in <1 kyr (Stern et al., 2003; Gainey and ElwoodMadden, 2012). Decomposition will release CH4 (Figs. 4-5). This CH4 isassumed in our model to be swiftly outgassed to the atmosphere, in partbecause of fracturing induced by the large volume change involved inclathrate decomposition.

The CH4 release to the atmosphere is proportional to the fraction ofsurface area initially shrouded by H2O-ice. This fraction is � 50%

3 Although the growing polar (H2O þ CO2)-ice cap overburden stabilizes polarclathrate, this represents <10% of the planet’s surface area and is omitted.

6

according to the model of Wordsworth et al. (2015).3 The initial CH4(g)flux is also proportional to f, the fraction of the not-yet-degassed HSZvolume occupied by clathrate. The clathrate is charged up with CH4produced by water-rock reactions over >108 yr (Fig. 2). It is released<103 yr after destabilization (Stern et al., 2003; Gainey and ElwoodMadden, 2012), possibly via explosive blow-outs or mud volcanism(Andreassen et al., 2017; Komatsu et al., 2016; Bro�z et al., 2019).

PCH4 on Mars is < 1 mbar during Step 2. As on today’s Mars, theoutgassing source is swamped by photochemical sinks during Step 2(Krasnopolsky et al., 2004). Moreover, what little CH4(g) exists is radia-tively ineffective during Step 2. This is because CH4–CO2collision-induced absorption (CIA) is weak when thecollapsed-atmosphere PCO2 is low (Wordsworth et al., 2017; Turbet et al.,2019).

Step 3. CO2 atmosphere re-inflation and further CH4 release. Mars at-mospheric collapse occurs at low obliquity (Soto et al., 2015). During theinterval of atmospheric collapse, PCO2 is set by vapor pressure equilib-rium with polar temperature. Most CO2 is sequestered as ice (Fig. A2).Obliquity can fall further after collapse is triggered, but obliquity willeventually rise. As obliquity rises, the collapsed atmosphere is still toothin to warm the poles – a hysteresis effect (Fig. A2). Although H2O iceshields CO2 ice from peak summer temperatures, H2O ice insulationraises annual-mean polar temperatures, which also increase with obliq-uity (Fig. 3). When obliquity rises to the point where no value of pressureyields a polar temperature below the condensation point, the CO2 icecaps are no longer stable. As a result, the remaining polar CO2 ice capsrapidly (103 yr) and completely sublimate (Fig. 6). As a result, the CO2greenhouse effect rapidly increases planet-wide.

Our scenario assumes that the CO2 in the ice caps is not irreversiblyentombed or otherwise irreversibly sequestered. However, if CO2 ice isdeposited on top of H2O ice, then it can be buried by gravitationalinstability for some parameter choices (Turbet et al., 2017). In our modelH2O ice is deposited on top of CO2 ice, which also acts to shield the CO2ice. Finally, basal melting can lead to sequestration of CO2 as liquid in thecrust. On the other hand, although stiff H2O ice may enshroud the CO2ice, flow of thick, soft CO2 ice (Mellon, 1996) opens crevasses, which


over time will allow CO2 to return to the atmosphere. There is no evi-dence for large subsurface reservoirs of liquid CO2. Therefore thisassumption, while not proven, is reasonable. We also neglect the adsor-bed CO2 reservoir in our calculations, because its estimated size (~35mbar or less; Jakosky, 2019) is much less than the ~1 bar needed for awarm climate in our model.

At this point, the low-latitudes have lost some or all of the H2O iceoverburden (much of it migrated away in Step 2). This H2O ice over-burden had the effect of stabilizing CH4 clathrate. H2O ice returns to thelow latitudes slowly: the wait time for all H2O ice to return to the lowlatitudes is > 106 yr. The atmosphere fully reinflates in a time muchshorter than this: ~103 yr. So, without the clathrate-stabilizing effect ofH2O ice overburden, clathrate is now exposed to the CO2 greenhousewarming associated with re-inflation. This CO2 greenhouse warming isstrong: for example, ~30 K for a 2 bar atm (Forget et al., 2013; Mischnaet al., 2013). As a result, clathrate irreversibly decomposes, releasingCH4. The wait time for CH4 clathrate decomposition is, for 102 m depth toHSZ, equal to (re-inflation time þ subsurface conductive-warming timeþ decomposition time)¼ (103 yrþ 102 yrþ<10 yr)� 103 yr. Therefore,CH4 is outgassed before H2O ice can re-load the highlands andre-stabilize the CH4 clathrate. The wait time for outgassing is also muchless than the CH4 photolysis timescale. Because of rapid release (Fig. A6),the atmospheric concentration of CH4 can be large after Step 3 (e.g.,Fig. 6).

The quantity of CH4 release depends on the duration of the atmo-spheric collapse, the prior unloading history of the CH4 clathrate reser-voir, and the volume of the shallow-subsurface CH4 clathrate reservoir.These have the following effect:

In the collapse-initiated CH4-release scenario, CH4 release is maximalfor the first atmospheric collapse in Mars’ history that lasts for ≳1Myr(i.e., a collapse so deep that the peaks of individual 105 yr obliquitycycles are not enough to trigger atmospheric re-inflation). This isbecause such a prolonged collapse gives time for substantialunloading of the CH4-stabilizing water ice overburden from thehighlands. For a prolonged collapse, much of the remaining CH4-clathrate reservoir is destabilized upon re-inflation. Inspection of anensemble of long-term obliquity integrations (Kite et al., 2015),combined with our GCM-derived collapse thresholds (Fig. 3), sug-gests that the first prolonged collapse happens geologically soon(≪109 yr) after Mars’ first-ever atmospheric collapse. By contrast, alarge CH4-burst does not result from <105 yr collapses. For such briefcollapses, unloading is so incomplete (according to the low subli-mation rates computed using our general circulationmodel) that mostof the highlands clathrate is still stabilized by the highlands H2O icethat is still in-place at the time of re-inflation.

In our model, once degassed, pore space is not recharged with CH4,because diffusion of CH4 through cold clathrate is slow. Thus, build-up of mbar of CH4 in the atmosphere by the collapse-initiated CH4release mechanism is only plausible during the first ≪1 Gyr of Mars’atmospheric-collapse history. (This point presumes that CH4 comesfrom deep hydrothermal alteration of ultramafic rock. However, ifpore space is filled with some amount of liquid H2O, olivine-hostedfluid inclusions can form abiotic CH4 from H2 produced via fluid-inclusion-scale serpentinization as well as radiolysis over relativelyshort (sub-Gyr) timescales.

The initial (pre-collapse) fraction of the hydrate stability zone that isoccupied by CH4 clathrate is unknown. We define f as the fraction ofthe not-yet-degassed portion of the HSZ that is initially occupied byclathrate. f/θ, where θ is porosity, gives the volume of pore space thatis occupied by clathrate.4 For f> 6%, the atmosphere contains 2–10%CH4 at the end of Step 3. This mix can allow CH4–CO2 collision

4 The value of f can also be interpreted as a measure of degassing efficiency ifsome CH4(g) is trapped by permafrost.

7

induced absorption (CIA) to provide climate warming. In order forCIA-warming to exceed the surface cooling caused by CH4 absorptionof sunlight, it is a requirement that the atmosphere has>(0.5–1.0) barCO2 (Turbet et al., 2019). Throughout this paper, we use the CH4–CO2Collision Induced Absorption warming appropriate to the experi-mental measurements of Turbet et al. (2019). (Qualitatively similarexperimental results are reported by Godin et al., 2019) We do thisbecause the warming calculations of Wordsworth et al. (2017) arebased on calculations, not experiments. If we instead adopt thewarming calculated by Wordsworth et al. (2017), then we wouldpredict warming that is about twice as strong (Fig. 6, Fig. A3).Therefore, our choice is conservative, in that it minimizes the effectsof CH4 bursts and the positive feedbacks of CH4 outgassing on furtherCH4 release.

Step 4. Possible warming due to CH4-enhanced greenhouse effect? Thecocktail of circumstances enabled by Mars’ first prolonged atmosphericcollapse (a massive CH4 pulse, a thick CO2 atmosphere, low-latitude H2Oice) can potentially warm Mars (Fig. 6, Fig. A3), but special circum-stances are required. H2O snow falling on the equator (plus any H2O icethat did not have time to sublimate) encounters high insolation (due tolower latitude) and potentially increases greenhouse forcing. The>255 Kmean annual temperature threshold, highlighted by the thick green linesin Fig. 6, is important because it is approximately the lower limit forseasonal meltwater runoff to form valleys that drain into perennial ice-covered lakes, as in modern Antarctica; e.g., McKay et al., 1985;McKay et al., 2005). In Fig. 6 the annual mean temperatures are shownfor the �40� latitude, �2 to þ3 km elevation zone (valley network zone;Hynek et al., 2010), which is warmer than the global average.

However, CH4-induced warming will only occur if atmosphericcollapse is initiated at >(0.5–1.0) bar pCO2. Currently available modelsof atmospheric collapse (Appendix A) do not show atmospheric collapseat such high values of pCO2 (for Early Mars luminosity). On the otherhand, collapse occurs at higher CO2 partial pressure if the surface ice/snow cover is extensive, raising significantly the planetary albedo ofEarly Mars (Ramirez, 2017). Moreover, other GCMs return different re-sults (Fig. A2), and the threshold pressure for atmospheric collapse issensitive to parameters (e.g., Kitzmann, 2016). Therefore, although thehypothesis of Early Mars warming due to atmospheric-collapse initiatedCH4 release is not ruled out by our study, it requires specialcircumstances.

We did photochemical calculations to find the lifetime of quickly-released CH4 (Appendix A). We found that CH4 can remain in the at-mosphere at radiatively important levels for 105–106 yr. If CH4 permitswarming, then this is long enough for frozen ground under lakes to un-freeze, linking lake water to deep aquifers. Such connections might helpto reconcile a generally cold Early Mars climate with evidence forgroundwater flow through fractures within sedimentary rocks on Mars(e.g., Yen et al., 2017; Frydenvang et al., 2017; Grasby et al., 2014; Wardand Pollard, 2018).

The end of the methane pulse: CH4 destruction (105-106 yr). Photo-chemical processes destroy atmospheric CH4 (Fig. A7). Eventually, theatmosphere will again collapse (Fig. 6). The newly collapsed atmospherecan retain many mbar of CH4. That is because at the temperature wherenearly all the CO2 has condensed, none of the atmospheric CH4 hascondensed. As a result, the atmosphere CH4/CO2 ratio can transientlyexceed the threshold for Titan-like photochemistry, haze, and soot(McKay et al., 1991; Haqq-Misra et al., 2008). This sooty phase is brief,because CH4 destruction is rapid when CH4/CO2 is high (Fig. A7). If allCH4 is converted to soot while CH4/CO2 > 0.1, then the soot layer col-umn mass can be> 100 kg/m2. This is sufficient to contaminate the crustat the organic matter levels reported by Eigenbrode et al. (2018) to adepth of ~7 km, or to 300 m depth for the ~0.1 wt% levels suggested byanalysis of the temperature spectrum of the SAM EGA (Sutter et al.,

Fig. 7. CH4 warming plot. ps ¼ partial pressure of atmospheric CO2 at Mars’ surface. Dashed lines correspond to warming calculated on the basis of experiments byTurbet et al. (2019), and solid lines correspond to warming calculated on the basis of calculations by Wordsworth et al. (2017). If the Wordsworth et al. (2017) resultsare used, then stronger CH4-induced climate warming would be obtained. Note that for ps ¼ 0.5 bar, adding CH4 can cause cooling. This is due to absorption ofnear-infrared insolation by CH4.

Fig. 8. Map of feedback strength (color shading corresponds to gain as definedin Eqn. (1)) for a single-column model, assuming initial surface temperature of240 K. Values of f > 0.3 exceed plausible porosity and so are unlikely.


2017).

4. Discussion

4.1. Sensitivity tests show that special conditions are needed for collapse-initiated CH4 release to cause warming

One key control for determining whether atmospheric-collapse-initiated CH4 release can cause surface warming is the fraction of HSZvolume occupied by clathrate, f (Fig. 8, Fig. A4). For f < 0.01, strongsurface warming is precluded. For f > 0.1, warming can be so strong thatsnowmelt can feed lakes. Methane is released by the warming triggeredby prior CH4 release. This feedback CH4 release increases with f. Thepositive feedback on CH4-induced warming can be large when f > 0.1(Fig. 8).

For the calculations underlying Fig. 8, we find the runaway clathratebreakdown threshold by comparing direct CH4-induced warming to thefeedback warming (induced by the CH4 release driven by direct warm-ing). If the indirect, feedback warming exceeds the direct warming, thenthe system will run away (i.e., all CH4 clathrate will be destroyed). Thisonly occurs for f > 0.3, which exceeds plausible porosity and so is un-likely. Otherwise the asymptotic warming gain (G) is given approxi-mately by

G ¼ 1/(1-R) (1)

(linearizing the feedback), where R is the feedback factor (Roe, 2009).Fig. A5 shows results including nonlinear feedbacks.

In our GCM, atmospheric collapse can be avoided (even for zeroobliquity) down to initial pressures <0.6 bar. This is unfavorable forcollapse-initiated CH4–CO2 climate warming. The collapse trigger at-mospheric pressure has to be >(0.5–1.0) bar, otherwise surface warmingfrom CH4 is weak or even net-negative (Turbet et al., 2019, Figs. 8 and 7).

CH4 release increases strongly with initial T. This is because of thenonlinearity of the CH4 clathrate decomposition curve (Fig. 4, Fig. A5).

The maximum size of a CH4 burst is reduced if the HSZ partiallyoutgasses prior to the main burst. Pre-warm-climate outgassing can occur

8

due to obliquity variations (Kite et al., 2017a), or as the result of iceunloading during pre-main-burst atmospheric collapses. We used anensemble of long-term obliquity simulations (Kite et al., 2015) to explorethese effects. In these calculations, we made the simplifying assumptionthat CH4 release was proportional to the volume of H2O ice removed, butwith CH4 release suppressed from depths that had previously beendegassed. This simplifying assumption allowed us to carry out >30long-term simulations. We determined the fraction of runs for which atleast 100 m of H2O ice first-time unloading occurs during a single at-mospheric collapse interval, and the dependence of this fraction on theobliquity for re-inflation and on the ice sublimation rate. The 100 mfirst-time ice-unloading value was chosen as a threshold for strong CH4outgassing. The “first-time” qualifier refers to the high probability that

Fig. 9. Warm paleoclimates indicated by geomorphology on Mars (modified after Kite et al., 2017b). Estimated ages are from Michael (2013). Y-axis corresponds tothe map-view scale of the landforms shown. Neither the durations of geologic eras, nor the durations of river-forming climates, are to scale. Data are consistent withlong, globally dry intervals.


the CH4-clathrate reservoir is not recharged by hydrothermal circulation.Because of the lack of recharge by this mechanism, repeated iceunloading will only yield CH4 gas from a given latitude-longitude-depthvolume element for the first time that volume element is unloaded(considering only the hydrothermal-circulation mechanism for CH4production; other mechanisms are possible, e.g. Klein et al., 2019). Wefound that the fraction of runs with a strong CH4 burst increases steeplywhen the difference between the obliquity for collapse and the obliquityfor re-inflation increases (i.e. when the “height” of the hysteresis loop inFig. A2 increases). However, the fraction of runs with a strong CH4 burstis insensitive to the ice unloading rate.

Consistent with the results of GCM simulations (e.g., Soto et al.,2015), the scenario assumes that CO2-atmosphere collapse occurs onlyfor φ < 40�. Otherwise, no major H2O ice shift will occur at collapse,because H2O ice is stable at low latitudes for low PCO2 when φ > 40�.

Overall, our conclusion is that special circumstances are required inorder to uncork 1017 kg of CH4, the minimum needed for strongwarming.

4.2. Predictions

If atmospheric collapse initiated a methane burst on Early Mars, thenthis would have consequences for Mars geology. These consequenceslead to testable predictions. For example, diameter<50 m impact cratersthat predate the methane burst should be very rare or absent (Kite et al.,2014; Warren et al., 2019; Vasavada et al., 1993). This is because if PCO2was high before the methane burst as required to avoid atmosphericcollapse, then small impactors would have been screened by the atmo-sphere, and so small hypervelocity impact craters would not form. Inaddition, pre-methane-burst large wind ripples (Lapotre et al., 2016)should be absent, because these only form at low atmospheric density.

Clathrate destabilization on Mars has been proposed to explain bothchaos terrain and also mounds interpreted as mud volcanoes (e.g., Mil-ton, 1974; Baker et al., 1991; Komatsu et al., 2000; Skinner and Tanaka,2007; Kite et al., 2007; Komatsu et al., 2016; Ivanov et al., 2014; Oehlerand Etiope, 2017; Pan and Ehlmann, 2014; Etiope and Oehler, 2019; Bro�zet al., 2019). Clathrate destabilization has also been proposed as a triggerfor the formation of chaos terrain.

If methane bursts occurred on Early Mars, that could also entail newinterpretations of Mars geochemistry. One possibility is deposition of>100 kg/m2 of photochemical soots. Such soots can indicate a reducingpast atmosphere (CH4/CO2 ≳ 0.1) (Trainer et al., 2006; Haqq-Misra et al.,2008). CH4/CO2 ≳ 0.1 is predicted to occur immediately following at-mospheric recollapse. Other mechanisms for rapidly delivering CH4 tothe Mars atmosphere, such as large-amplitude changes in mean obliquity(Kite et al., 2017a) or impact delivery (Haberle et al., 2018), might alsoproduce photochemical soots. This prediction is testable using theSHERLOC (Scanning Habitable Environments with Raman &

9

Luminescence for Organics & Chemicals) instrument on the Mars 2020rover (Beegle et al., 2015). Soots (complex abiotic organic matter) withinancient sediments are a potential life detection false positive (Sutteret al., 2017; Neveu et al., 2018; Eigenbrode et al., 2018).

4.3. Implications

Our collapse-initiated CH4-release scenario, which considers onlyCH4 produced by serpentinization, only works if serpentinizationoccurred on Mars (Lasue et al., 2015). However, the fraction of the crustthat must undergo serpentinization is small (<0.1%). Indeed, althoughserpentine is detected on Mars, these detections are overall uncommon(Ehlmann et al., 2010; Amador et al., 2018; Leask et al., 2018). This is notsurprising, because serpentinization may bemostly restricted to rocks toodeep for subsequent exhumation to the surface (Carter et al., 2013; Sunand Milliken, 2015). Moreover, radiolysis þ fluid-inclusion-scale ser-pentinization can produce H2þCH4 and not leave behind a mineralogicbyproduct that would be detectable by CRISM (Klein et al., 2019).Regardless of production depth and process, CH4-charged waters may beswept to the near-surface by hydrothermal circulation (Parmentier andZuber, 2007).

The collapse-initiated CH4-release scenario tends to produce, at most,one methane spike. If this occurred >0.5 Gyr after Mars formation, thenthe warm climate could be associated with valley network formation(Fig. 9). Previously proposed triggers for warm climates on Mars maysupplement the mechanism discussed here and are not inconsistent withit (e.g., Kite, 2019 and references therein).

In the collapse-initiated CH4 release scenario, a CH4-induced warmclimate (if it occurs at all) is preceded and postdated by a cold climatewith a thin atmosphere. This has broadly negative implications for Marssurface astrobiology. For astrobiology rovers, fluviolacustrine depositsascribed to the Noachian/Hesperian boundary represent candidatelanding sites with good potential for organic preservation and habit-ability (Summons et al., 2011; Ehlmann et al., 2008). Indeed, the delta inJezero crater is the planned landing site for the Mars 2020 rover (Fassettand Head, 2005; Ehlmann et al., 2008; Goudge et al., 2018). However, inthe collapse-initiated CH4-release scenario, Mars’ surface is sterilized(temperatures ~200 K and >1 MRad surface radiation doses; Hassleret al., 2014) just before the lakes are filled. The lakes might neverthelessbe inoculated with life from hot-spring or subsurface refugia.

5. Conclusions

We investigated a mechanism in which rapid andmassive (>30mbar)degassing of CH4-clathrate occurs consequent to atmospheric collapseand atmospheric reinflation. In this collapse-initiated CH4-release sce-nario, atmospheric collapse causes low-latitude surface water ice tosublimate away, depressurizing and thus destabilizing CH4 clathrate in


subglacial pore space. Subsequent atmospheric re-inflation leads towarming that further destabilizes CH4 clathrate, leading to more out-gassing. The CH4 bloom is brought to a close by photochemicaldestruction of CH4 (potentially accelerated by a new atmosphericcollapse). During the CH4 bloom, in some of our model runs, CH4/CO2transiently exceeds the threshold for haze and soot – a “Titan-like” blip inEarly Mars history. Drawdown of the CH4-clathrate reservoir ensures thatthe first CH4 spike is also the most intense.

Currently-available models indicate that special circumstances wouldhave been required for CH4-induced warming to occur and attain meanannual temperatures >255 K (consistent with perennial ice-coveredlakes; McKay et al., 2005). Specifically, the fraction of thenot-yet-degassed portion of the clathrate hydrate stability zone that isoccupied by clathrate (f) must exceed 0.1, and atmospheric collapse mustoccur for PCO2 > 0.8 bar. Our warming model output is also sensitive tothe choice of CH4–CO2 CIA opacities: experimentally derived parameters(Turbet et al., 2019) lead to colder outcomes than the parametersincluded in HITRAN (Wordsworth et al., 2017; Karman et al., 2019).

Declaration of competing interest

The authors declare that they have no known competing financialinterests or personal relationships that could have appeared to influencethe work reported in this paper.

10

CRediT authorship contribution statement

Edwin S. Kite: Conceptualization, Formal analysis, Funding acqui-sition, Investigation, Methodology, Project administration, Software,Visualization, Writing - original draft, Writing - review & editing.Michael A. Mischna: Investigation, Methodology, Software, Writing -review & editing. Peter Gao: Investigation, Methodology. Yuk L. Yung:Software. Martin Turbet: Investigation.

Acknowledgements

We thank Jesse Tarnas, and an anonymous reviewer, for helpful re-views that led to an improved manuscript. We thank Alan Howard, ColinGoldblatt, Ross Irwin, Bob Craddock, Alejandro Soto, John Armstrong,Feng Tian, Itay Halevy, Alex Pavlov, Tom McCollom, Curtis Manning,Sarah Stewart, and Chris Oze for discussions, and Robin Wordsworth forsharing model output. The MATLAB scripts and GCM summary outputused to make the figures in this paper may be obtained for unrestrictedfurther use by contacting the lead author. This project has receivedfunding from the European Union’s Horizon 2020 research and innova-tion program under the Marie Skłodowska-Curie Grant Agreement No.832738/ESCAPE. M.T. and E.S.K. acknowledge support from the FranceAnd Chicago Collaborating in The Sciences (FACCTS) program. E.S.K.acknowledges funding from NASA (NNX16AG55G).

Appendix A. Methods

A.1. Overview

In order to model CH4 release initiated by atmospheric collapse and subsequent atmospheric reinflation, we use simulated obliquity forcing to drivea model of surface temperature evolution (Fig. A1). Temperature is calculated as a function of latitude, longitude, and depth beneath the surface. Themodel uses a GCM-derived parameterization of atmospheric collapse and re-inflation (Fig. A2). For the surface temperature boundary condition, a GCM-derived look-up table is used. Changes in surface temperature and overburden pressure cause subsurface CH4-clathrate to decompose and outgas. Onceoutgassed, CH4 can be destroyed by photochemical processes. CH4 can either warm or cool the surface, depending on PCO2. Different assumptions aboutlong-term escape-to-space of volatiles can be incorporated by varying initial conditions. Once initialized, the total surface-exchangeable H2O and CO2

inventories (i.e., sum of the atmosphere þ ice caps þ shallow ground ice reservoirs) are held constant for the duration of a model run (107–108 yr). Ineffect, we assume that neither escape-to-space of H2O and CO2, nor geologic sequestration of H2O and CO2, cause a large fractional change in the H2Oand CO2 inventories during the model run (107–108 yr).

The total CH4 outgassed due to Mars atmospheric collapse and re-inflation can be decomposed into:

Prompt CH4 release from atmospheric collapse (reduction in PCO2, combined with planetwide cooling). This is always zero, because the largestabilizing effect of cooling defeats the small destabilizing effect of the loss of the weight of the atmosphere.

Minor CH4 release from H2O ice unloading following atmospheric collapse (Step 2 in Fig. 2). Major, re-inflation CH4 release: from warming during re-inflation (Steps 3–4 in Fig. 2). CH4 release (if any) due to warming feedbacks (Steps 3–4 in Fig. 2).

Direct CH4 release is proportional to the fraction of the not-yet-degassed portion of the HSZ that is initially occupied by clathrate (f). Feedback CH4release increases nonlinearly with f.

Fig. A1. Schematic of the model. HSZ ¼ CH4-clathrate Hydrate Stability Zone.


A.2. Atmospheric Collapse and Surface Temperature Modeling

To calculate atmospheric collapse as a function of obliquity and PCO2 (Fig. A2), we used the Mars Weather Research and Forecasting general cir-culation model (MarsWRF GCM; Mischna et al., 2013; Richardson et al., 2007; Toigo et al., 2012) to build a look-up table for Mars annual-mean surfacetemperatures as a function of obliquity, log of PCO2, and latitude, for a solar luminosity 75% of modern (corresponding to 0.7 Ga after Mars formationaccording to the standard solar model; Bahcall et al., 2001). Four PCO2 values were investigated: 6 mbar, 60 mbar, 600 mbar, and 1200 mbar. Fiveobliquities were investigated 0�, 5�, 15�, 25� and 35�. Other values of PCO2 and obliquity were obtained by extrapolation and interpolation from this gridof 5 � 4 ¼ 20 runs. In each scenario, we began with a dry atmosphere, and a water ice north polar cap ranging from 1 to 10,000 kg/m2 thicknessdepending on the simulation. There is no CO2 ice on the surface to begin with. We assumed a spatially and temporally uniform atmospheric dust opacityof 0.3. This is a modest amount of dust, and ignores the complicating factors of seasonality and strength that are unknowns for different obliquities andsurface pressures. For comparison, we also investigated a second scenario with zero dust. All cases contain radiatively active H2O and CO2 ice cloudswhere saturation is reached. We used a polynomial fit to these results to determine the boundary of atmospheric collapse (i.e., coldest temperatures atthe CO2 condensation point). Radiatively active CO2 ice aerosol is included. CO2 ice grain size radius is assumed to be 100 μm. This choice is based onthe argument of Forget et al. (1998) that these larger particles would be appropriate in a CO2-rich atmosphere. Results are likely sensitive to thisassumption (Kitzmann, 2016). The GCM runs used for this study use a dust scenario that is derived fromMars Global Surveyor data for a year without aplanet-encircling dust storm and used in the LMD Mars Climate Database. In the vertical, a Conrath-ν dust profile is used.

Collapse rate is set to ~1 mbar/yr (Soto et al., 2015). We use the same rate for re-inflation (Soto et al., 2015). We force the model with φ(t) fromrealistic obliquity histories. We vary φ, but leave the eccentricity and the longitude of perihelion set to present-day values. We do this because φ is thedominant control on polar insolation (Schorghofer, 2008). Collapse takes ~1 kyr and is followed by a collapsed interval of 104–107 yr duration and thenby ~1 kyr of re-inflation (Fig. 6). We adopt φ c ¼ 16.5�.

Fig. A2. Atmospheric-collapse phase portraits for a collapse-triggering obliquity of ~16.5�, for two different climate models. Results from the GCM of Mischna et al.(2013), with dust, are shown in orange. Loop fit to output of the GCM of Forget et al. (2013) shown in yellow. The hysteresis loops that are highlighted by thick linesare examples of climate system trajectories for a total CO2 inventory of 1.66 bar (orange) and 0.74 bar (yellow) – different conditions are required for the differentmodels. The part of the orange curve that closes the loop from arrow 2 to arrow 3 occurs at very low atmospheric pressure (<0.1 mbar). The thin lines correspond topressure-temperature combinations on the polar–CO2–condensation curve that are unstable on 102–103 yr (CO2 condensation) timescales, so are not physicallyrealizable. The high-PCO2, low-obliquity vertical “stalk” in the thick orange line corresponds to atmospheres that stay inflated due to CH4–CO2 collision-inducedabsorption (CIA). These solutions are stable on 102–103 yr (CO2 condensation) timescales, but decay on 105–106 yr (CH4 photolysis) timescales.

A.3. Clathrate modeling and CH4–CO2 warming

The collapse-initiated CH4-release scenario assumes the existence of shallow CH4-clathrate on Mars, with CH4 produced abiotically in associationwith deep water-rock reactions (Kite et al., 2017a; Etiope and Sherwood Lollar, 2013; Mousis et al., 2013). The P-T pathway forming CH4-clathrate isshown in Fig. 4.

Guided by the GCM output, we calculated T and P as a function of depth within the regolith, latitude, and longitude, using Tsurf from the GCMs, andassumptions about H2O ice distribution motivated by the water-cycle output of previously published GCMs (e.g., Mischna et al., 2013; Wordsworthet al., 2015). Spatial resolution is (2� latitude) � (2� longitude) � (1 m depth). Tsurf decreases with elevation, more so as PCO2 increases (Wordsworth,2016). The initial H2O ice distribution is set by allocating 1.4 � 107 km3 of H2O ice (Mahaffy et al., 2015) uniformly across land with elevations>1 km(using modern topography). Where H2O ice rests on top of the regolith, we model the corresponding overpressure and thermal insulation (assumingthermal conductivity, kice ¼ 2 W m�1 K�1). The grid is truncated at 300 m depth-within-regolith. This is because destabilization of clathrate to depths

11


>300m requires annual-mean Tsurf> 273 K, much warmer than the climate model. For 300m, the vertical thermal conduction timescale is 1 kyr –muchshorter than the other timescales affecting atmospheric PCH4. In particular, the CH4 destruction timescale (when the atmosphere is inflated) is 105–106

yr. This separation of timescales holds even when the low thermal diffusivity of CH4 clathrate and the latent heat of clathrate decomposition areconsidered. Therefore, we approximate geotherm adjustment as rapid.

Assuming rapid geotherm adjustment, we calculate the depth-to-HSZ. We express this depth as meters below the top of the regolith, so a depth-to-HSZ of 100m below an ice sheet of 200m thickness means that clathrate is stable�300m below the surface. We use the phase diagram of Sloan and Koh(2008, their Table 4.1), as shown in Fig. 4. We set porosity¼ 0.3 (high values of porosity are supported by rover gravimetry; Lewis et al., 2019), assume120 kg CH4/(m3 clathrate), lithospheric heat flow Q ¼ 0.03 W m�2, regolith thermal conductivity kreg ¼ 2.5 W m�1 K�1, and regolith density ρreg ¼2000 kg m�3. For the purpose of modeling the effect of H2O ice overburden on CH4-clathrate stability, we consider both a case with H2O ice on highground and a case without H2O ice on high ground. The resulting “depth to Hydrate Stability Zone” look-up tables are passed to the main driver.

The main driver uses the following inputs. (1) The depth-to-HSZ look-up tables (Fig. 5). (2) The phase portraits for atmospheric collapse from theGCMs (Fig. A2). (3) A CH4 destruction look-up table (described below) (Fig. A7). (4) A parameterization of greenhouse warming due to CH4 -CO2collision-induced absorption (CIA) that is based on the experimental results reported in Turbet et al. (2019) (Fig. A4). (5) A simulated obliquity timeseries obtained using the N-body code of Chambers (1999) and an obliquity wrapper script (Armstrong et al., 2004); see Kite et al. (2015) for details. Tosave computer time, we shift the obliquity time series up and down as needed to trigger collapse relatively early in the run (the likely true wait times foratmospheric collapse depend on the initial obliquity of Mars, and range from only a few Myr if Mars’ obliquity was initially low, to hundreds of Myr ifMars’ obliquity was initially high).

Fed by these inputs, the main driver first calculates the depth-to-HSZ for four key states (shown in Fig. 5), as follows: (1) pre-collapse state (H2O iceon highlands, inflated atmosphere – before Step 1 in Fig. 2); (2) immediately post-collapse (H2O ice on highlands, collapsed atmosphere – end of Step 1in Fig. 2); (3) fully unloaded but collapsed state (negligible H2O ice on highlands, collapsed atmosphere – end of Step 2 in Fig. 2); and (4) immediatelyafter re-inflation (negligible H2O ice on highlands, re-inflated atmosphere – Step 3 in Fig. 2). For some atmospheric collapses, the duration of collapse istoo brief for complete unloading of H2O ice from the highlands. Therefore, we computed compute depth-to-HSZ following re-inflation for a range oftime intervals that are too short for complete unloading of H2O ice. We assume that unloading is steady and uniform (we explored nonuniformunloading specifications, but found only trivial differences). Next, based on a HSZ occupancy fraction, f, we compute the volume of CH4 clathrate thatdecomposes on CO2-atmosphere re-inflation for complete H2O-ice unloading of the highlands. Methane decomposition is rapid (Stern et al., 2003;Gainey and Elwood Madden, 2012). CH4 clathrate breakdown involves a >14% reduction in solid volume, and we assume fractures allow methane gasreleased at �100 m depth to reach the surface in≪105 yr. The greenhouse warming (if any) corresponding to this CH4 outgassing is calculated (Fig. 7,Turbet et al., 2019). The corresponding maximum CH4 release, including CH4-warming-induced CH4 release, is obtained by iteration. For incompleteH2O ice unloading of the highlands, we linearize both the direct CH4 release in response to re-inflation, and also the CH4-warming-induced feedbackCH4 release. Linearization is invalid for the most extreme warming (see Fig. A5), but is reasonable for determining whether or not a warm climate canoccur.

Fig. A3. Example warming map showing geographic distribution of rivers and lakes, at peak warming, for the optimistic climate evolution shown in Fig. 6. Specialcircumstances are required for conditions this warm. Colored contours show topography in meters (present-day topography is used). Before CH4 release, conditionswarmer than lakeshore weather stations in Taylor Valley, Antarctica only occur for the lowest elevations (inside thick black contour line), far from mapped valleynetworks (light gray dots; Hynek et al., 2010). However, with 17 K of uniform CH4-induced warming (Fig. 6), the area warmer than lakeshore weather stations inTaylor Valley Antarctica expands to cover (thick red line) most of the planet (45�S-85�N). Dark gray area is obscured by young lavas.

For large f, clathrate decomposition and outgassing can runaway (Fig. A5). Runaway outgassing can produce a Mars climate (Fig. A5) with Tave >

273 K – a temperate climate (Halevy et al., 2011; Bishop et al., 2018). A temperate Mars climate lasting ~1Myr might explain the surface leaching eventapparently recorded by aluminous clays overlying Fe/Mg clays at many globally-distributed locations (Carter et al., 2015; Bishop et al., 2018). Runawayclathrate breakdown depends on f, the pre-release temperature, and PCO2. For cooler pre-release temperatures, the minimum f for runaway clathratebreakdown increases, due to nonlinearity in the clathrate phase diagram (Fig. A5).

12


Fig. A4. Climate evolution model output. As Fig. 6, but showing sensitivity to a reduction in f. Solid lines are for f ¼ 0.06, dashed lines are for the value (f ¼ 0.15) usedin Fig. 6. For f ¼ 0.06, the CH4-induced warm climate lasts about half as long, and the peak CH4-boosted annual-average temperatures in the valley network zone donot exceed 253K (horizontal black line in lower panel). Because of these lower peak temperatures, collapse triggered methane outgassing for f ¼ 0.06 is unlikely toexplain Mars’ valley networks.

Fig. A5. Showing the (narrow) range of conditions under which CH4 outgassing on Mars can lead to runaway warming and outgassing from an f ¼ 0.15 clathratereservoir. Colored dashed lines correspond to CH4-induced warming (Turbet et al., 2019). Solid lines show the corresponding CH4 release. The arrows labeled 1–3 givean example of how to read the diagram. Supposing (1) an initial collapsed initiated CH4 release of 30 mbar CH4 in a 1000 mbar PCO2 atmosphere, the warming (2) is~4K. For an initial surface temperature Tinit ¼ 220K, this is sufficient to release (3) a further 18 mbar of CH4, which will lead to further warming and thus more CH4

release. When the warming (dashed line) for a given PCO2 plots above a CH4 release line (solid line), runaway outgassing occurs. When the warming for a given PCO2plots below a CH4 release line, there is no runaway, but positive feedback still occurs. These results are for a simplified model using a single column ofclathrate-charged regolith.

The time-stepping loop in the main driver uses the phase portraits in Fig. A2, as well as warming due to CH4 (if any), to track atmospheric collapseand re-inflation. Water ice net sublimation rate, i, is set to i ¼ 0.15 mm/yr for migration from high ground to poles. This low value (based on MarsWRFGCM results) is due to the low vapor pressure of H2O in the ~200 K collapsed atmosphere. Movement of ice from poles back to high ground under thewarm re-inflated atmosphere should be faster; we use i ¼ �1.5 mm/yr. Our results are qualitatively unaffected by reasonable changes in i. Theoverpressure due to polar ice caps is not explicitly modeled. This is acceptable because polar cap area is small, and, as a result, the fraction of clathrateprotected by the polar cap overburden is, likewise, small. For collapse duration Δt� (~300m)/i, the H2O ice is completely removed from the highlands.If Δt< (~300m)/i, we use a look-up table to find the partial CH4 release. We also track release of CH4 from previous re-inflations and collapses to ensurethat the same portion of the regolith cannot release CH4 twice (Fig. A6). Finally, we track inhibition of orbitally paced atmospheric collapses by CH4

warming, if any warming occurs (Fig. 6).

13

Planetary and Space Science 181 (2020) 104820
E.S. Kite et al.
Fig. A6. Methane inventory history for one run with f ¼ 0.15. We track release of CH4 from previous re-inflations and collapses to ensure that the same portion of theregolith cannot release CH4 twice. .

Relative to climate models that include CO2 and H2O alone, the amount of warming needed to bring Early Mars valley network locations to tem-peratures sufficiently high for rivers and lakes is 10 K–20 K. For warming exceeding 10K–20K, temperatures over most of the Mars Southern Highlandsexceed measured temperatures in Taylor Valley Antarctica on the shores of ice covered perennial lakes (Fig. A3) (Doran et al., 2002). Lakes overspill toform regionally integrated valley networks. In Antarctica, a melt-fed lake-overspill river incises into bedrock (Shaw and Healy, 1980). High lake-bottomtemperatures destabilize sub-lake CH4 clathrates (Fig. 5). Another possible positive feedback, not included in our model, is H2 production by aqueousweathering of olivine at Mars’ surface during warm climates (Tosca et al., 2018). H2 is a strong greenhouse gas via H2–CO2 CIA (Ramirez et al., 2014;Ramirez, 2017; Turbet et al., 2019).

A.4. Methane destruction parameterization

We used the Caltech/JPL 1-D Mars photochemistry code, modified to include CH4 (and C2H6, and C2H2, etc.) (Summer et al. 2002; Nair et al. 1994,2005; Nair et al., 1994), to build a look-up table for CH4 destruction by photochemical processes. The CH4 destruction model is the same as that in Kiteet al. (2017a), except for the use of a UV flux appropriate for the Sun 3.8 Ga (Claire et al., 2012). Boundary conditions include surface burial of O2, O3,H2O2, and CO.Water vapor is set to saturation at the surface; results are insensitive to H2O concentration. As in Kite et al. (2017), we found that the maincontrol on CH4 destruction rate is the CH4/CO2 ratio, so we fit a curve to the CH4 destruction rate as a function of the CH4/CO2 ratio for use in theclimate-evolution model (Fig. A7).

Fig. A7. CH4 destruction rate. Both calculations adopt a UV flux appropriate for the Sun 3.8 Ga (Claire et al., 2012), and a high H2O vapor mixing ratio. Because a highH2O vapor mixing ratio tends to speed the destruction of CH4 in our photochemical calculations, our use of a high H2O vapor mixing ratio is conservative in that it willtend to understate the duration of CH4-induced warming. .

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.pss.2019.104820.
References
Amador, E.S., Bandfield, J.L., Thomas, N.H., 2018. A search for minerals associated withserpentinization across Mars using CRISM spectral data. Icarus 311, 113–134.

14

Andreassen, K., et al., 2017. Massive blow-out craters formed by hydrate-controlledmethane expulsion from the Arctic seafloor. Science 356 (6341), 948–953. https://doi.org/10.1126/science.aal4500.


http://refhub.elsevier.com/S0032-0633(19)30270-3/sref1



https://doi.org/10.1126/science.aal4500

https://doi.org/10.1126/science.aal4500


Armstrong, J.C., Leovy, C.B., Quinn, T., 2004. A 1 Gyr climate model for Mars: neworbital statistics and the importance of seasonally resolved polar processes. Icarus171, 255–271.

Bahcall, J.N., Pinsonneault, M.H., Basu, S., 2001. Solar models: current epoch and timedependences, neutrinos, and helioseismological properties. Astrophys. J. 555,990–1012.

Baker, V.R., Strom, R.G., Gulick, V.C., Kargel, J.S., Komatsu, G., 1991. Ancient oceans, icesheets and the hydrological cycle on Mars. Nature 352, 589–594.

Batalha, N.E., Kopparapu, Ravi Kumar, Haqq-MisraJacob, Kasting, J.F., 2016. Climatecycling on early Mars caused by the carbonate-silicate cycle. Earth Planet. Sci. Lett.455, 7–13.

Beegle, L., et al., 2015. SHERLOC: scanning habitable environments with Raman &luminescence for organics & chemicals, 2015. Aero. Conf. IEEE. https://doi.org/10.1109/AERO.2015.7119105.

Bierson, C.J., Phillips, R.J., Smith, I.B., Wood, S.E., Putzig, N.E., Nunes, D., Byrne, S.,2016. Stratigraphy and evolution of the buried CO2 deposit in the Martian southpolar cap. Geophys. Res. Lett. 43 (9), 4172–4179.

Bishop, J.L., Fair�en, A.G., Michalski, J.R., Gago-Duport, L., Baker, L.L., Velbel, M.A.,Gross, C., Rampe, E.B., 2018. Surface clay formation during short-term warmer andwetter conditions on a largely cold ancient Mars. Nat. Astron. 2, 206–213.

Bowen, G.J., Maibauer, B.J., Kraus, M.J., R€ohl, U., Westerhold, T., Steimke, A., 2016. Twomassive, rapid releases of carbon during the onset of the Palaeocene-Eocene thermalmaximum. Nat. Geosci. 8, 44–47.

Bro�z, P., Hauber, E., van de Burgt, I., �Spillar, V., Michael, G., 2019. Subsurface sedimentmobilization in the southern chryse planitia on Mars. https://doi.org/10.1029/2018JE005868.

Carter, J., Poulet, F., Bibring, J.-P., Mangold, N., Murchie, S., 2013. Hydrous minerals onMars as seen by the CRISM and OMEGA imaging spectrometers: updated global view.J. Geophys. Res.: Planets 118, 831–858.

Carter, J., Loizeau, D., Mangold, N., Poulet, F., Bibring, J.-P., 2015. Widespread surfaceweathering on early Mars: a case for a warmer and wetter climate. Icarus 248,373–382.

Cassata, W.S., Shuster, D.L., Renne, P.R., Weiss, B.P., 2012. Trapped Ar isotopes inmeteorite ALH 84001 indicate Mars did not have a thick ancient atmosphere. Icarus221, 461–465.

Chambers, J.E., 1999. A hybrid symplectic integrator that permits close encountersbetween massive bodies. Mon. Not. R. Astron. Soc. 304, 793–799.

Chassefi�ere, E., Leblanc, F., 2011. Methane release and the carbon cycle on Mars. Planet.Space Sci. 59, 207–217.

Chassefi�ere, E., Lasue, J., Langlais, B., Quesnel, Y., 2016. Early Mars serpentinizationderived CH4 reservoirs and H2 induced warming. Meteorit. Planet. Sci. 51,2234–2245.

Chastain, B.K., Chevrier, V., 2007. Methane clathrate hydrates as a potential source formartian atmospheric methane. Planet. Space Sci. 55, 1246–1256.

Claire, M.W., et al., 2012. The evolution of solar flux from 0.1 nm to 160 μm. Astrophys.J. 757, 12 article id. 95.

Doran, P.T., et al., 2002. Valley floor climate observations from the McMurdo dry valleys,Antarctica, 1986-2000. J. Geophys. Res. Atmos. 107, D24. https://doi.org/10.1029/2001JD002045. ACL 13-1, CiteID 4772.

Ehlmann, B.L., Mustard, J.F., Fassett, C.I., Schon, S.C., Head III, J.W., Des Marais, D.J.,Grant, J.A., Murchie, S.L., 2008. Clay minerals in delta deposits and organicpreservation potential on Mars. Nat. Geosci. 1 (6), 355–358.

Ehlmann, B.L., Mustard, J.F., Murchie, S.L., 2010. Geologic setting of serpentine depositson Mars. Geophys. Res. Lett. 370, L06201. https://doi.org/10.1029/2010GL042596,2010.

Ehlmann, B.L., et al., 2011. Subsurface water and clay mineral formation during the earlyhistory of Mars. Nature 479 (7371), 53–60.

Eigenbrode, J.L., Summons, R.E., Steele, A., Freissinet, C., Millan, M., Navarro-Gonz�alez, R., Sutter, B., McAdam, A.C., Franz, H.B., Glavin, D.P., Archer Jr., P.D.,Mahaffy, P.R., Conrad, P.G., Hurowitz, J.A., et al., 2018. Organic matter preserved in3-billion-year-old mudstones at Gale crater, Mars. Science 360 (6393), 1096–1101.

Etiope, Giuseppe, Oehler, Dorothy Z., 2019. Methane spikes, background seasonality andnon-detections on Mars: a geological perspective. Planet. Space Sci. 168, 52–61.

Etiope, G., Schoell, M., 2014. Abiotic gas: atypical, but not rare. Elements 10 (4),291–296.

Etiope, G., Sherwood Lollar, B., 2013. Abiotic methane on Earth. Rev. Geophys. 51,276–299.

Fassett, C.I., Head, J.W., 2005. Fluvial sedimentary deposits on Mars: ancient deltas in acrater lake in the Nili Fossae region. Geophys. Res. Lett. 32 (14), L14201.

Fassett, C.I., Head, J.W., 2011. Sequence and timing of conditions on early Mars. Icarus211, 1204–1214. https://doi.org/10.1016/j.icarus.2010.11.014.

Fastook, J.L., Head III, J.W., 2015. Glaciation in the Late Noachian Icy Highlands: iceaccumulation, distribution, flow rates, basal melting, and top-down melting rates andpatterns. Planet. Space Sci. 106, 82–98. https://doi.org/10.1016/j.pss.2014.11.028.

Forget, F., Hourdin, F., Talagrand, O., 1998. CO2 snowfall on Mars: simulation with ageneral circulation model. Icarus 131, 302–316.

Forget, F., et al., 2013. 3D modelling of the early martian climate under a denser CO2atmosphere: temperatures and CO2 ice clouds. Icarus 222 (1), 81–99.

Frydenvang, J., Gasda, P.J., Hurowitz, J.A., Grotzinger, J.P., Wiens, R.C., Newsom, H.E.,Edgett, K.S., Watkins, J., Bridges, J.C., Maurice, S., Fisk, M.R., Johnson, J.R.,Rapin, W., Stein, N.T., Clegg, S.M., Schwenzer, S.P., Bedford, C.C., Edwards, P.,Mangold, N., Cousin, A., Anderson, R.B., Payr�e, V., Vaniman, D., Blake, D.F.,Lanza, N.L., Gupta, S., Van Beek, J., Sautter, V., Meslin, P.-Y., Rice, M., Milliken, R.,Gellert, R., Thompson, L., Clark, B.C., Sumner, D.Y., Fraeman, A.A., Kinch, K.M.,Madsen, M.B., Mitrofanov, I.G., Jun, I., Calef, F., Vasavada, A.R., 2017. Diagenetic

15

silica enrichment and late-stage groundwater activity in Gale crater. Mars, Geophys.Res. Lett. 44 (10), 4716–4724. https://doi.org/10.1002/2017GL073323.

Gainey, S.R., Elwood Madden, M.E., 2012. Kinetics of methane clathrate formation anddissociation under Mars relevant conditions. Icarus 218, 513–524.

Gierasch, P.J., Toon, O.B., 1973. Atmospheric pressure variation and the climate of Mars.J. Atmos. Sci. 30, 1502–1508.

Godin, Paul J., Ramirez, Ramses M., Campbell, Charissa, Wizenberg, Tyler, GiangNguyen, Tue, Strong, Kimberly, Moores, John E., 2019. Far infrared measurements ofabsorptions by CH4þ CO2 and H2þ CO2 mixtures and implications for greenhousewarming on early Mars. ESSOAr. https://doi.org/10.1002/essoar.10501475.1.

Goudge, T.A., Mohrig, D., Cardenas, B.T., Hughes, Cory M., Fassett, C.I., 2018.Stratigraphy and paleohydrology of delta channel deposits, Jezero crater, Mars.Icarus 301, 58–75.

Grasby, S.E., Proemse, B.C., Beauchamp, B., 2014. Deep groundwater circulation throughthe High Arctic cryosphere forms Mars-like gullies. Geology 42, 651–654.

Grotzinger, J., Sumner, D.Y., Kah, L.C., Stack, K., Gupta, S., Edgar, L., Rubin, D., Lewis, K.,Schieber, J., Mangold, N., Milliken, R., Conrad, G., DesMarais, D., Farmer, J.,Siebach, K., Calef, F., Hurowitz, J., McLennan, S.M., et al., 2014. A Habitable Fluvio-lacustrine environment at Yellowknife bay, Gale crater, Mars. Science 343 (6169),1242777.

Haberle, R.M., Catling, D.C., Carr, M.H., Zahnle, K.J., 2017. The early Mars climatesystem. In: Haberle, R.M., et al. (Eds.), The Atmosphere and Climate of Mars.Cambridge University Press, pp. 497–525. ISBN: 9781139060172.

Haberle, R.M., Zahnle, K., Barlow, N.G., 2018. Warming Early Mars by Impact Degassingof Reduced Greenhouse Gases, 49th Lunar and Planetary Science Conference 19-23March, 2018 held at The Woodlands, Texas LPI Contribution No. 2083, id.1682.

Halevy, I., Head, J.W., 2017. In: Atmospheric Mass and the Geologic Record of Water onMars, 48th Lunar and Planetary Science Conference, Held 20-24 March 2017 at TheWoodlands, Texas. LPI Contribution No. 1964, id.2519.

Halevy, I., Fischer, W.W., Eiler, J.M., 2011. Carbonates in the Martian meteorite AllanHills 84001 formed at 18�4�C in a near-surface aqueous environment. Proc. Natl.Acad. Sci. 108, 16895–16899.

Haqq-Misra, J.D., Domagal-Goldman, S.D., Kasting, P.J., Kasting, J.F., 2008. A revised,hazy methane greenhouse for the archean Earth. Astrobiology 8, 1127–1137.

Hassler, D.M., et al., 2014. Mars’ surface radiation environment measured with the Marsscience laboratory’s curiosity rover. Science 343, 1244797.

Hecht, M.H., 2002. Metastability of liquid water on Mars. Icarus 156, 373–386.Heng, K., Kopparla, P., 2012. On the stability of Super-Earth atmospheres. Astrophys. J.

754, 1.Holo, S.J., Kite, E.S., Robbins, S.J., 2018. Mars obliquity history constrained by elliptic

crater orientations. Earth Planet. Sci. Lett. 496, 206–214.H€orst, S.M., 2017. Titan’s atmosphere and climate. J. Geophys. Res.: Planets 122 (3),

432–482.Hynek, B., 2016. The great climate paradox of ancient Mars. Geology 44, 879–880.Hynek, B.M., Beach, M., Hoke, M.R.T., 2010. Updated global map of Martian valley

networks and implications for climate and hydrologic processes. J. Geophys. Res. 115(E9). CiteID E09008.

Irwin, R.P., Lewis, K.W., Howard, A.D., Grant, J.A., 2015. Paleohydrology of Eberswaldecrater, Mars. Geomorphology. https://doi.org/10.1016/j.geomorph.2014.10.012.

Isherwood, R.J., Jozwiak, L.M., Jansen, J.C., Andrews-Hanna, J.C., 2013. The volcanichistory of Olympus Mons from paleo-topography and flexural modeling. Earth Planet.Sci. Lett. 363, 88–96.

Ivanov, M.A., Hiesinger, H., Erkeling, G., Reiss, D., 2014. Mud volcanism and morphologyof impact craters in Utopia Planitia on Mars: evidence for the ancient ocean. Icarus228, 121–140.

Jakosky, B.M., 2019. The CO2 inventory on Mars. Planet. Space Sci. 175, 52–59.Jakosky, B.M., Edwards, C.S., 2018. Inventory of CO2 available for terraforming Mars.

Nature Astronomy 2, 634–639.Jakosky, B.M., et al., 2018. Loss of the Martian atmosphere to space: present-day loss

rates determined from MAVEN observations and integrated loss through time. Icarus315, 146–157.

Kahre, M.A., Vines, S.K., Haberle, R.M., Hollingsworth, J.L., 2013. The early Martianatmosphere: investigating the role of the dust cycle in the possible maintenance oftwo stable climate states. J. Geophys. Res. – Planets. https://doi.org/10.1002/jgre.20099.

Karman, Tijs, Gordon, Iouli E., van der Avoird, Ad, Baranov, Yury I., Boulet, Christian,Drouin, Brian J., Groenenboom, Gerrit C., Gustafsson, Magnus, Hartmann, Jean-Michel, Kurucz, Robert L., Rothman, Laurence S., Sun, Kang, Sung, Keeyoon,Thalman, Ryan, Tran, Ha, Wishnow, Edward H., Wordsworth, Robin, Vigasin, AndreyA., Volkamer, Rainer, van der Zande, Wim, J., 2019. Update of the HITRAN collision-induced absorption section. Icarus 328, 160–175.

Kite, E.S., 2019. Geologic constraints on early Mars climate. Space Sci. Rev. 215, 10.Kite, Edwin S., Hindmarsh, Richard C.A., 2007. Did ice streams shape the largest channels

on Mars? Geophys. Res. Lett. 34 (19). CiteID L19202.Kite, E.S., Hovius, N., Hillier, J.K., Besserer, J., 2007. Candidate Mud Volcanoes in the

Northern Plains of Mars. American Geophysical Union. Fall Meeting 2007, abstractid.V13B-1346.

Kite, E.S., Williams, J.-P., Lucas, A., Aharonson, O., 2014. Low palaeopressure of themartian atmosphere estimated from the size distribution of ancient craters. Nat.Geosci. 7, 335–339.

Kite, E.S., Howard, A., Lucas, A., Armstrong, J.C., Aharonson, O., Lamb, M.P., 2015.Stratigraphy of Aeolis Dorsa, Mars: stratigraphic context of the great river deposits.Icarus 253, 223–242.

Kite, E.S., Gao, P., Goldblatt, C., Mischna, M., Mayer, D., Yung, Y.L., 2017a. Methanebursts as a trigger for intermittent lake-forming climates on post-Noachian Mars. Nat.Geosci. 10, 737–740.












http://refhub.elsevier.com/S0032-0633(19)30270-3/optcu5NG8dDMK




https://doi.org/10.1109/AERO.2015.7119105

https://doi.org/10.1109/AERO.2015.7119105
















https://doi.org/10.1029/2018JE005868

https://doi.org/10.1029/2018JE005868

































https://doi.org/10.1029/2001JD002045

https://doi.org/10.1029/2001JD002045





https://doi.org/10.1029/2010GL042596





















https://doi.org/10.1016/j.icarus.2010.11.014

https://doi.org/10.1016/j.pss.2014.11.028








https://doi.org/10.1002/2017GL073323







https://doi.org/10.1002/essoar.10501475.1


















































https://doi.org/10.1016/j.geomorph.2014.10.012


















https://doi.org/10.1002/jgre.20099

https://doi.org/10.1002/jgre.20099



























Kite, E.S., Sneed, J., Mayer, D.P., Wilson, S.A., 2017b. Persistent or repeated surfacehabitability on Mars during the late Hesperian - Amazonian. Geophys. Res. Lett.https://doi.org/10.1002/2017GL072660.

Kite, E.S., Mayer, D.P., Wilson, S.A., Davis, J.M., Lucas, A.S., Stucky de Quay, G., 2019.Persistence of intense, climate-driven runoff late in Mars history. Sci. Adv. 5 (3)https://doi.org/10.1126/sciadv.aav7710 eaav7710.

Kurahashi-Nakamura, T., Tajika, E., 2006. Atmospheric collapse and transport of carbondioxide into the subsurface on early Mars. Geophys. Res. Lett. 33, 18. https://doi.org/10.1029/2006GL027170.

Kitzmann, D., 2016. Revisiting the scattering greenhouse effect of CO2 ice clouds.Astrophys. J. Lett. 817, 5 article id. L18.

Klein, F., Grozeva, N.G., Seewald, J.S., 2019. Abiotic methane synthesis andserpentinization in olivine-hosted fluid inclusions. Proc. Natl. Acad. Sci. 116,17666–17672.

Komatsu, G., Kargel, J.S., Baker, V.R., Strom, R.G., Ori, G.G., Mosangini, C., Tanaka, K.L.,2000. A chaotic terrain formation hypothesis: explosive outgas and outlow bydissociation of clathrate on Mars. In: 31st Annual Lunar and Planetary ScienceConference, March 13-17, 2000. Houston, Texas, abstract no. 1434.

Komatsu, G., Okubo, C.H., Wray, J.J., Ojha, L., Cardinale, M., Murana, A., Orosei, R.,Chan, M.A., Orm€o, J., Gallagher, R., 2016. Small edifice features in Chryse Planitia,Mars: assessment of a mud volcano hypothesis. Icarus 268, 56–75.

Korablev, Oleg, Vandaele, Ann Carine, Montmessin, Franck, Fedorova, Anna A.,Trokhimovskiy, Alexander, Forget, François, Lef�evre, Franck, Daerden, Frank,Thomas, Ian R., Trompet, Loïc, Erwin, Justin T., Aoki, Shohei, Robert, S�everine,Neary, Lori, Viscardy, S�ebastien, Grigoriev, Alexey V., Ignatiev, Nikolay I.,Shakun, Alexey, Patrakeev, Andrey, Belyaev, Denis A., Bertaux, Jean-Loup,Olsen, Kevin S., Baggio, Lucio, Alday, Juan, Ivanov, Yuriy S., Ristic, Bojan,Mason, Jon, Willame, Yannick, Depiesse, C�edric, Hetey, Laszlo, Berkenbosch, Sophie,Clairquin, Roland, Queirolo, Claudio, Beeckman, Bram, Neefs, Eddy, Patel, ManishR., Bellucci, Giancarlo, L�opez-Moreno, Jose-Juan, Wilson, Colin F., Etiope, Giuseppe,Zelenyi, Lev, Svedhem, Hakan, Vago, Jorge L., Acs; NOMAD Science Teams, 2019. Nodetection of methane on Mars from early ExoMars trace gas orbiter observations.Nature 568 (7753), 517–520.

Krasnopolsky, V.A., Maillard, J.P., Owen, T.C., 2004. Detection of methane in the martianatmosphere: evidence for life? Icarus 172, 537–547.

Kreslavsky, M.A., Head, J.W., 2005. Mars at very low obliquity: atmospheric collapse andthe fate of volatiles. Geophys. Res. Lett. 32 (12). CiteID L12202.

Kurokawa, H., Kurosawa, K., Usui, T., 2017. A lower limit of atmospheric pressure onEarly Mars inferred from nitrogen and argon isotopic compositions. Icarus. https://doi.org/10.1016/j.icarus.2017.08.020.

Lammer, H.E., et al., 2013. Outgassing history and escape of the martian atmosphere andwater inventory. Space Sci. Rev. 174, 113–154.

Lapotre, M.G.A., et al., 2016. Large wind ripples on Mars: a record of atmosphericevolution. Science 353, 55–58.

Laskar, J., et al., 2004. Long term evolution and chaotic diffusion of the insolationquantities of Mars. Icarus 170, 343–364.

Lasue, J., Quesnel, Y., Langlais, B., Chassefi�ere, E., 2015. Methane storage capacity of theearly martian cryosphere. Icarus 260, 205–214.

Leask, E.K., Ehlmann, B.L., Dundar, M.M., Murchie, S.L., Seelos, F.P., 2018. Challenges inthe search for Perchlorate and other hydrated minerals with 2.1-μm absorptions onMars. Geophys. Res. Lett. https://doi.org/10.1029/2018GL080077, 09 November2018.

Lewis, Kevin W., Peters, Stephen, Gonter, Kurt, Morrison, Shaunna, Schmerr, Nicholas,Vasavada, Ashwin R., Gabriel, Travis, 2019. A surface gravity traverse on Marsindicates low bedrock density at Gale crater. Science 363 (6426), 535–537.

Lin, Li-Hung, Hall, James, Lippmann-Pipke, Johanna, Ward, Julie A., SherwoodLollar, Barbara, Deflaun, Mary, Rothmel, Randi, Moser, Duane, Gihring, Thomas M.,Mislowack, Bianca, Onstott, T.C., 2005. Radiolytic H2 in continental crust: nuclearpower for deep subsurface microbial communities. Geochem. Geophys. Geosyst. 6(7), Q07003.

Lunine, Jonathan I., Atreya, Sushil K., 2008. The methane cycle on Titan. Nat. Geosci. 1,159–164.

Luo, W., Cang, X., Howard, A.D., 2017. New Martian valley network volume estimateconsistent with ancient ocean and warm and wet climate. Nat. Commun. 8, 15766.

Lyons, J.R., Manning, C., Nimmo, F., 2005. Formation of methane on Mars by fluid-rockinteraction in the crust. Geophys. Res. Lett. 32. CiteID L13201.

Mahaffy, P.R., et al., 2015. The imprint of atmospheric evolution in the D/H of Hesperianclay minerals on Mars. Science 347, 412–414.

Manning, C.V., McKay, C.P., Zahnle, K.J., 2006. Thick and thin models of the evolution ofcarbon dioxide on Mars. Icarus 180, 38–59.

Manning, C.V., Bierson, C., Putzig, N.E., McKay, C.P., 2019. The formation and stability ofburied polar CO2 deposits on Mars. Icarus 317, 509–517.

McKay, C.P., Wharton Jr., R.A., Squyres, S.W., Clow, G.D., 1985. Thickness of ice onperennially frozen lakes. Nature 313, 561.

McKay, C.P., Pollack, J.B., Courtin, R., 1991. The greenhouse and antigreenhouse effectson Titan. Science 253, 1118–1121.

McKay, C.P., Andersen, D.T., Pollard, W.H., Heldmann, J.L., Doran, P.T., Fritsen, C.H.,Priscu, J.C., 2005. Polar lakes, streams, and springs as analogs for the hydrologicalcycle on Mars. In: Water on Mars and Life. Springer, Berlin, Heidelberg, pp. 219–233.

Mellon, M.T., 1996. Limits on the CO2 content of the martian polar deposits. Icarus 124,268–279.

Michael, G.G., 2013. Planetary surface dating from crater size-frequency distributionmeasurements: Multiple resurfacing episodes and differential isochron fitting. Icarus226 (1), 885–890.

Milton, D.J., 1974. Carbon dioxide hydrate and floods on Mars. Science 183, 654–656.

16

Mischna, Michael A., Allen, Mark, Richardson, Mark I., Newman, Claire E.,Toigo, Anthony D., 2011. Atmospheric modeling of Mars methane surface releases.Planet. Space Sci. 59, 227–237.

Mischna, M.A., Baker, V., Milliken, R., Richardson, M., Lee, C., 2013. Effects of obliquityand water vapor/trace gas greenhouses in the early martian climate. J. Geophys. Res.118, 560–576.

Moores, J.E., et al., 2019. The methane diurnal variation and micro-seepage flux at Galecrater, Mars as constrained by the ExoMars Trace Gas Orbiter and Curiosityobservations. Geophys. Res. Lett. https://doi.org/10.1029/2019GL083800.

Mousis, O., et al., 2013. Volatile trapping in martian clathrates. Space Sci. Rev. 174,213–250.

Nair, H., Allen, M., Anbar, A.D., Yung, Y.L., Clancy, R.T., 1994. A photochemical model ofthe martian atmosphere. Icarus 111, 124–150.

Nair, H., Summers, M.E., Miller, C.E., Yung, Y.L., 2005. Isotopic fractionation of methanein the martian atmosphere. Icarus 175, 32–35.

Nakamura, T., Tajika, E., 2001. Stability and evolution of the climate system of Mars.Earth Planets Space 53, 851–859.

Nakamura, T., Tajika, E., 2002. Stability of the Martian climate system under the seasonalchange condition of solar radiation. J. Geophys. Res. 107 (E11), 4-1, CiteID 5094.

Nakamura, T., Tajika, E., 2003. Climate change of Mars-like planets due to obliquityvariations: implications for Mars. Geophys. Res. Lett. 30, 18-1, CiteID 1685.

Neveu, Marc, Hays, Lindsay E., Voytek, Mary A., New, Michael H., Schulte, Mitchell D.,2018. The ladder of life detection. Astrobiology 18, 1375–1402.

Oehler, Dorothy Z., Allen, Carlton C., 2010. Evidence for pervasive mud volcanism inacidalia Planitia, Mars. Icarus 208 (2), 636–657.

Oehler, Dorothy Z., Etiope, Giuseppe, 2017. Methane seepage on Mars: where to look andwhy. Astrobiology 17 (12), 1233–1264, 2017.

Oze, C., Sharma, M., 2005. Have olivine, will gas: serpentinization and the abiogenicproduction of methane on Mars. Geophys. Res. Lett. 32, 10. CiteID L10203.

Oze, C., Jones, L.C., Goldsmith, J.I., Rosenbauer, R.J., 2012. Differentiating biotic fromabiotic methane genesis in hydrothermally active planetary surfaces. Proc. Natl.Acad. Sci. 109, 9750–9754.

Pan, L., Ehlmann, B.L., 2014. Phyllosilicate and hydrated silica detections in the knobbyterrains of Acidalia Planitia, northern plains, Mars. Geophys. Res. Lett. 41,1890–1898.

Parmentier, E.M., Zuber, M.T., 2007. Early evolution of Mars with mantle compositionalstratification or hydrothermal crustal cooling. JGR (Planets) 112, E02007.

Phillips, R.J., Davis, B.J., Tanaka, K.L., et al., 2011. Massive CO2 ice deposits sequesteredin the South polar layered deposits of Mars. Science 332, 838–841.

Prieto-Ballesteros, O., et al., 2006. Interglacial clathrate destabilization on Mars: possiblecontributing source of its atmospheric methane. Geology 34, 149.

Putzig, Nathaniel E., Smith, Isaac B., Perry, Matthew R., Foss, Frederick J.,Campbell, Bruce A., Phillips, Roger J., Seu, Roberto, 2018. Three-dimensional radarimaging of structures and craters in the Martian polar caps. Icarus 308, 138–147.

Ramirez, R.M., 2017. A warmer and wetter solution for early Mars and the challengeswith transient warming. Icarus 297, 71–82.

Ramirez, R.M., Kopparapu, R., Zugger, M.E., Robinson, T.D., Freedman, R., Kasting, J.F.,2014. Warming early Mars with CO2 and H2. Nat. Geosci. 7, 59–63.

Richardson, M.I., Toigo, A.D., Newman, C.E., 2007. PlanetWRF: a general purpose, localto global numerical model for planetary atmospheric and climate dynamics.J. Geophys. Res. 112, E09001.

Roe, G., 2009. Feedbacks, timescales, and seeing red. Annu. Rev. Earth Planet Sci. 37,93–115.

Sagan, C., 1977. Reducing greenhouses and the temperature history of Earth and Mars.Nature 269, 224–226. https://doi.org/10.1038/269224a0.

Sagan, C., Toon, O.B., Gierasch, P., 1973. Climatic change on Mars. Science 181,1045–1049.

Schorghofer, N., 2008. Temperature response of Mars to milankovitch cycles. Geophys.Res. Lett. 35, L18201.

Shaw, J., Healy, T., 1980. Morphology of the Onyx River system, McMurdo sound region,Antarctica. N. Z. J. Geol. Geophys. 23 (2), 223–238.

Skinner, J.A., Tanaka, K.L., 2007. Evidence for and implications of sedimentary diapirismand mud volcanism in the southern Utopia highland–lowland boundary plain, Mars.Icarus 186, 41–59.

Sloan, E.D., Koh, C.A., 2008. Clathrate Hydrates of Natural Gases, third ed. CRC Press.Soto, A., Mischna, M., Schneider, T., Lee, C., Richardson, M.I., 2015. Martian atmospheric

collapse: idealized GCM studies. Icarus 250, 553–569.Steakley, K., Kahre, M., Haberle, R., Zahnle, K., 2019. In: EPSC/DPS Joint Meeting

abstract number EPSC-DPS2019-509-1.Stern, L.A., Circone, S., Kirby, S.H., Durham, W.B., 2003. Temperature, pressure, and

compositional effects on anomalous or ’’self’’ preservation of gas hydrates. Can. J.Phys. 81, 271–283.

Summons, R.E., et al., 2011. Preservation of martian organic and environmental records:final report of the Mars biosignature working group. Astrobiology 11 (2). https://doi.org/10.1089/ast.2010.0506.

Sun, V.Z., Milliken, R.E., 2015. Ancient and recent clay formation on Mars as revealedfrom a global survey of hydrous minerals in crater central peaks. J. Geophys. Res.Planets 120, 2293–2332.

Sutter, B., McAdam, A.C., Mahaffy, P.R., Ming, D.W., Edgett, K.S., Rampe, E.B.,Eigenbrode, J.L., Franz, H.B., Freissinet, C., Grotzinger, J.P., et al., 2017. Evolved gasanalyses of sedimentary rocks and eolian sediment in Gale Crater, Mars: results of theCuriosity rover’s sample analysis at Mars instrument from Yellowknife Bay to theNamib Dune. J. Geophys. Res.: Planets 122, 2574–2609.

Tarnas, J.D., Mustard, J.F., Sherwood Lollar, B., Bramble, M.S., Cannon, K.M.,Palumbo, A.M., Plesa, A.-C., 2018. Radiolytic H2 production on Noachian Mars:

https://doi.org/10.1002/2017GL072660

https://doi.org/10.1126/sciadv.aav7710

https://doi.org/10.1029/2006GL027170

https://doi.org/10.1029/2006GL027170

























































https://doi.org/10.1029/2018GL080077








































http://refhub.elsevier.com/S0032-0633(19)30270-3/opt5L4uOKeTyK














https://doi.org/10.1029/2019GL083800






























































https://doi.org/10.1038/269224a0
























https://doi.org/10.1089/ast.2010.0506

https://doi.org/10.1089/ast.2010.0506














implications for habitability and atmospheric warming. Earth Planet. Sci. Lett. 502,133–145.

Tarnas, J.D., Mustard, J.F., Sherwood Lollar, B., Cannon, K.M., Palumbo, A.M., Plesa, A.-C., Bramble, M.S., 2019. In: 50th Lunar and Planetary Science Conference, Held 18-22 March, 2019 at the Woodlands, Texas. LPI Contribution No. 2132, id.2029.

Tobie, G., Lunine, J.I., Sotin, C., 2006. Episodic outgassing as the origin of atmosphericmethane on Titan. Nature 440, 61–64.

Toigo, A.D., Lee, C., Newman, C.E., Richardson, M.I., 2012. The impact of resolution onthe dynamics of the martian global atmosphere: varying resolution studies with theMarsWRF GCM. Icarus 221, 276–288.

Tosca, N.J., Ahmed, Imad A.M., Tutolo, B.M., Ashpitel, A., Hurowitz, J.A., 2018.Magnetite authigenesis and the warming of early Mars. Nat. Geosci. 11, 635–639.

Touma, J., Wisdom, J., 1993. The chaotic obliquity of Mars. Science 259, 1294–1297.ISSN 0036-8075.

Trainer, M.G., Pavlov, A.A., Dewitt, H.L., Jimenez, J.L., McKay, C.P., Toon, O.B.,Tolbert, M.A., 2006. Inaugural article: organic haze on titan and the early Earth. Proc.Natl. Acad. Sci. 103, 18035–18042.

Turbet, M., Forget, F., Leconte, J., Charnay, B., Tobie, G., 2017. CO2 condensation is aserious limit to the deglaciation of Earth-like planets. Earth Planet. Sci. Lett. 476,11–21.

Turbet, M., Tran, H., Pirali, O., Forget, F., Boulet, C., Hartmann, J.-M., 2019. Far infraredmeasurements of absorptions by CH4þCO2 and H2þCO2 mixtures and implicationsfor greenhouse warming on early Mars. Icarus 321, 189–199.

Vasavada, A.R., 2017. Our changing view of Mars. Phys. Today 70 (3), 34–41.Vasavada, A.R., Milavec, T.J., Paige, D.A., 1993. Microcraters on Mars - evidence for past

climate variations. J. Geophys. Res. 98 (E2), 3469–3476. ISSN 0148-0227.Ward, M.K., Pollard, W.H., 2018. A hydrohalite spring deposit in the Canadian high

Arctic: a potential Mars analogue. Earth Planet. Sci. Lett. 504, 126–138.

17

Warren, A.O., Kite, E.S., Williams, J.-P., Horgan, B., 2019. Through the thick and thin:new constraints on Martian paleopressure history 3.8-4 Ga from small exhumedcraters. J. Geophys. Res. Planets. https://doi.org/10.1029/2019JE006178.

Webster, C.R., Mahaffy, P.R., Atreya, S.K., Moores, J.E., Flesch, G.J., Malespin, C.,McKay, C.P., Martinez, G., Smith, C.L., Martin-Torres, J., et al., 2018. Backgroundlevels of methane in Mars’ atmosphere show strong seasonal variations. Science 360,1093–1096.

Wordsworth, R., 2015. Atmospheric heat redistribution and collapse on tidally lockedrocky planets. Astrophys. J. 806 (2), 15 article id. 180.

Wordsworth, R., 2016. The climate of early Mars. Annu. Rev. Earth Planet Sci. 44,381–408. https://doi.org/10.1146/annurev-earth-060115-012355.

Wordsworth, R., Forget, F., Millour, E., Head, J.W., Madeleine, J.-B., Charnay, B., 2013.Global modelling of the early martian climate under a denser CO2 atmosphere: watercycle and ice evolution. Icarus 222, 1–19.

Wordsworth, R., Kerber, L., Pierrehumbert, R., Forget, F., Head, J.W., 2015. Comparisonof warm and wet and cold and icy scenarios for early Mars in a 3-D climate model.JGR: Planets 120, 1201–1219.

Wordsworth, R., et al., 2017. Transient reducing greenhouse warming on early Mars.Geophys. Res. Lett. 44, 665–671.

Wray, James J., Ehlmann, Bethany L., 2011. Geology of possible Martian methane sourceregions. Planet. Space Sci. 59 (2–3), 196–202.

Yen, A.S., Ming, D.W., Vaniman, D.T., Gellert, R., Blake, D.F., Morris, R.V.,Morrison, S.M., Bristow, T.F., Chipera, S.J., Edgett, K.S., Treiman, A.H., Clark, B.C.,Downs, R.T., Farmer, J.D., et al., 2017. Multiple stages of aqueous alteration alongfractures in mudstone and sandstone strata in Gale Crater, Mars. Earth Planet. Sci.Lett. 471, 186–198.

Zahnle, K., Catling, D., 2019. The paradox of Mars methane. In: Ninth InternationalConference on Mars abstract #6132.

Zahnle, K., Freedman, R.S., Catling, D.C., 2011. Is there methane on Mars? Icarus 212,493–503.













































https://doi.org/10.1029/2019JE006178








https://doi.org/10.1146/annurev-earth-060115-012355



























Planetary and Space Science - UChicago...

Documents

Transcript of Planetary and Space Science - UChicago...